Search

A systematic study was made of methods of increasing the light collection efficiency of scintillation counters. Various reflectors, surface treatment of the scintillator, relative geometries of scintillator and photodiode, and light pipes were tried, using plastic scintillator cubes, 3 and 6 in. on a side, and right circular cylinders, 3 in. in diameter by 3 in. in length and 6 in. in diameter by 6 in. in length. The maximum factors by which light output could be increased, relative to a polished scintillator, were about 2.8 for 6‐in. scintillators and 3.3 for 3‐in. scintillators without the use of light pipes, and 5.0 for a 3‐in. scintillator with light pipe. The fraction of light trapped in a right circular cylinder by total internal reflection is derived as a function of index of refraction, as are the percentages of light coming out the ends and sides of a cylinder.

For interactions of the type M1 + M2 ? M3 + M4 the kinematics of both initial particles and one secondary particle determine the reaction completely. Given a fixed initial state and any missing mass M4, the momentum-angle correlation for particle M3 is determined uniquely. In the spectrometer described, this correlation is preserved as a line image in the focal plane of a double focusing magnetic lens. Dispersions both in the plane of production and at right angles to it separate the images corresponding to different values of M4, for which the spectrum is obtained directly by means of a 28-counter hodoscope in the focal plane. Each hodoscope counter is gated with a scintillation and Cerenkov counter telescope to identify particle M3. During each beam pulse the outputs from the hodoscope counters are recorded on 10 Mc/sec scalers. Between accelerator pulses the contents of these scalers are read onto magnetic tape and into a PDP-5 computer, which accumulates, corrects, and displays the data continuously to provide an on-line monitor.